mne-tools/mne-python
Mne Python
MNE-Python is an open-source Python library for processing, visualizing, and analyzing human neurophysiological data, including MEG, EEG, sEEG, ECoG, and NIRS recordings. It provides a comprehensive framework for loading data from over 30 proprietary file formats into a common hierarchical FIF data structure, and represents all time-series data as NumPy arrays for seamless integration with the scientific Python ecosystem. The library is built around object-oriented data containers that encapsulate raw, epoched, evoked, and source data with built-in preprocessing and visualization methods.
The project distinguishes itself through a complete forward model and inverse solver pipeline for source localization, combining boundary element method forward models with distributed, beamformer, and dipole fitting inverse solvers. It includes a time-generalization decoding framework for training and testing classifiers across all time-point pairs to reveal temporal dynamics of neural representations, and offers cluster-based permutation testing to control family-wise error rates in neuroimaging statistics. The interactive Matplotlib visualization layer generates publication-quality static and interactive plots with specialized neuroimaging layouts and 3D brain surface rendering.
Beyond its core identity, MNE-Python supports functional connectivity analysis for quantifying statistical dependencies between brain regions, time-frequency decomposition of neural signals, and machine learning tools including common spatial pattern filters, receptive field modeling, and source space decoding pattern projection. The library also provides signal preprocessing capabilities for artifact detection and repair using regression, ICA, and signal-space separation, as well as neural data simulation for testing analysis pipelines. Statistical testing includes parametric and non-parametric permutation tests with spatio-temporal clustering applied to sensor-level, source-level, and source estimate data.
The software can be installed via pip or conda for users familiar with Python package management, or through a standalone installer that bundles all dependencies for users new to the environment.
Features
- MEG/EEG Data Analyses - Provides a comprehensive framework for loading, preprocessing, visualizing, and analyzing MEG, EEG, sEEG, ECoG, and NIRS recordings.
- Brain Source Localizers - Reconstructs brain activity locations from sensor measurements using inverse modeling.
- Neurophysiological Raw Data Manipulators - Queries, manipulates, annotates, plots, and exports continuous sensor recordings in the Raw data structure.
- Neurophysiological Data Visualizations - Creates interactive and static plots of sensor and source-level neurophysiological data for pattern exploration.
- Neurophysiological Signal Explorers - Loads, processes, and visualizes MEG, EEG, sEEG, ECoG, and NIRS recordings for scientific analysis.
- Neuroimaging Analysis Object Persistence - Loads and saves epochs, evoked responses, forward solutions, source estimates, and other analysis objects from disk.
- Event-Locked Epoch Segmenters - Divides continuous recordings into time-locked segments for event-related analysis and exports them to Pandas DataFrames.
- Custom File Format Readers - Loads data from over 30 proprietary formats through a modular plugin-based reader system.
- Time-Frequency Analysis - Decomposes neural signals into their frequency components over time to study oscillatory activity.
- Neural Signal Decompositions - Computes frequency and time-frequency representations of neural signals, including steady-state evoked potentials.
- Neuroimaging Visualization Extensions - Generates publication-quality static and interactive plots with specialized neuroimaging layouts and 3D brain surface rendering.
- Physiological Artifact Removers - Provides regression, ICA, SSP, and signal-space separation for artifact removal from neural recordings.
- Cortical Source Reconstruction Methods - Reconstructs brain activity at the source level using distributed, sparse, mixed-norm, beamformer, and dipole fitting methods.
- Neurophysiological Format Importers - Reads raw neurophysiological data from over 30 file formats including MEG, EEG, and fNIRS systems.
- Brain Connectivity Analyses - Implements spectral and effective connectivity measures to quantify statistical dependencies between brain regions.
- Brain Source Localization Frameworks - Ships a complete forward model and inverse solver pipeline for source localization using distributed, beamformer, and dipole fitting methods.
- Neuroimaging Source Localization Solvers - Ships a complete forward model and inverse solver pipeline for reconstructing brain activity from sensor measurements.
- Event-Related Potential Estimators - Averages epochs across experimental conditions to produce event-related potentials or fields with visualization.
- MEG/EEG Data Analysis Libraries - Provides an open-source Python library for loading, preprocessing, visualizing, and analyzing MEG, EEG, sEEG, ECoG, and NIRS data.
- Neuroimaging Statistical Testing Suites - Provides permutation tests, cluster statistics, and ANOVA on sensor and source-level neuroimaging data with multiple comparison correction.
- Neurophysiological Signal Processing Toolkits - Provides a comprehensive toolkit for filtering, artifact removal, and preprocessing of continuous and epoched neural recordings.
- Statistical Significance Testing - Runs parametric and non-parametric permutation tests with clustering for neuroimaging significance testing.
- NumPy Array Integration - Represents all time-series data as NumPy arrays for seamless integration with the scientific Python ecosystem.
- Spatio-Temporal Cluster Permutation Tests - Performs permutation t-tests with spatio-temporal clustering on source estimates.
- Spatio-Temporal Cluster Permutations - Provides cluster-based permutation testing to control family-wise error rates in neuroimaging statistics.
- Two-Sample Spatio-Temporal Cluster Permutation Tests - Compares two independent groups of source estimates with permutation tests and cluster correction.
- Neurophysiological Signal Processors - Loads, preprocesses, and analyzes MEG, EEG, sEEG, and ECoG data for research and clinical insights.
- Neurophysiological Signal Cleaners - Filters and cleans raw neurophysiological data to prepare it for downstream analysis.
- Neurophysiological Data Containers - Provides object-oriented data containers that encapsulate raw, epoched, evoked, and source data with built-in methods.
- Brain Source Localization Techniques - Reconstructs the location and time course of brain activity from sensor measurements using source estimation techniques.
- Spatio-Temporal Cluster ANOVAs - Applies repeated measures ANOVA with spatio-temporal clustering to source estimates.
- Cortical Source Space Constructors - Constructs cortical source spaces and leadfield matrices using boundary element models.
- Hierarchical Neurophysiological Data Formats - Defines a hierarchical FIF file format that preserves channel metadata, annotations, and processing history for neurophysiological data.
- Source Space Decoding Projections - Projects sensor-space decoding patterns onto the cortical surface for anatomical interpretation.
- Mixing Matrix Extractors - Computes mixing matrices from spatial filters to interpret neural signal composition.
- Neurophysiological Decoding Model Applications - Trains and evaluates advanced decoding models, including time generalization, on neurophysiological data.
- Spectro-Temporal Receptive Field Models - Implements linear encoding models to recover spectro-temporal receptive fields from neural recordings.
- Laplacian Smoothness Priors - Applies Laplacian smoothness priors to regularize receptive field estimates in neural data.
- Time-Resolved Classifiers - Fits classifiers at each time point of epoched data to reveal temporal discriminability.
- Cross-Temporal Decoding Frameworks - Implements a time-generalization decoding framework that trains and tests classifiers across all time-point pairs.
- Brain Data Decoding Models - Trains and evaluates decoding models on neurophysiological features for classification and regression tasks.
- Neural Data Decoding Modules - Provides a module for training decoding models, applying time generalization, and projecting classifier patterns to source space.
- Neural Receptive Field Estimators - Provides a module for fitting receptive field models to neural data with smoothness priors.
- Multitaper Spectral Feature Extractors - Ships a multitaper-based PSD feature extraction method for decoding neural data.
- Scalp Topography Maps - Generates topographic maps of the scalp field distribution at specified time points, with configurable averaging durations.
- Neuroimaging Joint Plots - Combines butterfly plots with automatically placed scalp topographies for a comprehensive overview of evoked data.
- Neurophysiological Data Exports - Writes epochs, evoked responses, and raw data to common file formats for sharing or further processing.
- Functional Brain Connectivity Metrics - Calculates all-to-all spectral and effective connectivity metrics to quantify interactions between brain regions.
- Topographical Sensor Layouts - Ships topographical subplot layouts that arrange per-sensor traces according to their spatial positions on the scalp.
- Neurophysiological Signal Traces - Provides specialized butterfly-style evoked response plotting with channel-type grouping and global field power overlay.
- Neurophysiological Field Map Plots - Estimates and visualizes the magnetic or electric field on the scalp or MEG helmet surface in a three-dimensional view.
- Neurophysiological Data Object Constructors - Constructs Raw, Epochs, and Evoked objects directly from NumPy arrays with user-defined channel information.
- Clinical Neurophysiology Processors - Processes sEEG, ECoG, and polysomnography data for clinical applications such as sleep staging.
- Common Spatial Pattern Filters - Implements common spatial pattern filters that maximize variance differences between two classes to discriminate neural activity.
- Cross-Temporal Decoding Generalizations - Trains classifiers at one time point and tests at all others to reveal temporal dynamics of neural representations.
- Functional Connectivity Analysis Tools - Computes spectral and effective connectivity measures to quantify statistical dependencies between brain regions.
- Evoked Response Comparison Plots - Plots multiple evoked conditions on the same axes, supporting various channel combination methods and dictionary-key-based styling.
- Neural Encoding Model Fittings - Estimates receptive fields from neural data with optional smoothness priors to model stimulus-response relationships.
- Neuroimaging Publication Graphics - Produces high-quality, interactive figures from neurophysiological data for reports and presentations.
- Neural Signal Dependency Quantifiers - Quantifies statistical dependencies between brain regions from neurophysiological signals to study functional networks.
- Neurophysiological Channel Image Plots - Displays an image plot with one channel per row, showing the evoked response across time for selected channels.
- Neurophysiological Arrow Overlays - Overlays arrows on a scalp topography to show the estimated magnitude and direction of the magnetic field at a given time.
- Neuroscience Programming - Processing time-resolved neural signals like EEG and MEG.
- Python Toolboxes - Comprehensive library for processing and visualizing neuroimaging data.
